The facts of the breast cancer epidemic are well known. In 1995, in the U.S. 182,000 women will succumb to this disease; 46,000 will die from it. Mortality from breast cancer has remained stable over the past 20 years. Systemic adjuvant therapies will reduce the annual odds of death from breast cancer by 25%. The contributions of newer drugs or altered sequences of drug administration to these figures are as yet unclear. For patients with advance disease the outlook is uniformly grim. Despite initial responses to systemic therapy, the majority of the patients will suffer progressive disease within 1 year of diagnosis and be dead by 3. This is due to the proliferation of drug-resistant cells, present from diagnosis in all patients with metastatic disease and the majority with node-positive cancer. Extreme intensification with high dose chemotherapy (HDC) will improve response rates in women with metastatic disease, resulting in a small proportion of long term survivors, and may substantially augment at least shortterm disease-free survival in the high-risk adjuvant setting. However, despite intensification of therapy, the majority of women with metastatic disease will relapse. We have investigated the induction of autologous graft-vs-host disease (AGVHD) after marrow transplantation (BMT) as a potential immune mechanism for eradicating micrometastatic disease and have previously shown that treatment with cyclosporine A (CSA) will induce a syndrome of AGVHD in both animals and humans, that the syndrome is associated with an anti-tumor effect directed against tumor targets which bear the MHC Class II (HLA-Dr) antigen, and that the syndrome can be intensified by co-administration of gamma interferon with CSA. In addition, these preliminary studies suggest the induction of AGVHD will reduce the risk of progression after HDC in women with metastatic breast cancer. We now propose to further investigate critical clinical and mechanistic questions about the induction of AGVHD as a form of anti-cancer immunotherapy after BMT for women with breast cancer. We will perform 2 clinical trials and address the following 4 aims: 1. Determine if AGVHD can be induced in women whose high-dose therapy is supported by PBPC and bone marrow, or PBPC alone. 2. Determine the toxicities associated with IL-2 administration following HDC and stem cell infusion in women undergoing induction of AGVHD, and specifically, find an optimal dose for further study. 3. Determine if the in vitro antitumor effect associated with CsA-induced AGVHD can be enhanced by both IL-2 and by strategies which augment tumor cell recognition. 4. Determine if the T cell receptor (TcR) repertoire of AGVHD reactive lymphocytes is limited.